Run Brief
Define objective, allowed scope, hard stops, required inputs, expected output, and verification before work starts.
HTML preview gallery for AI coding-agent operators
Templates for proving agent work is actually done.
Browse the public proof layer for the $29 Agent Ops Command Center pack: run logs, done checks, cost audits, routing decisions, handoffs, delivery proof, and buyer approval assets.
Start And Accept Runs
Use these before and after an agent edits code, docs, releases, or buyer-facing surfaces.
Run Audit
Review objective match, scope match, proof, cost, handoff, security, and revenue claims before accepting output.
Done Checklist
Prevent false completion with objective, scope, regression, evidence, handoff, safety, cost, and revenue gates.
Acceptance Test Suite
Turn agent output into pass, fail, or needs-review decisions using concrete proof signals.
Verification Command Library
Map code, UI, release, secret, rollback, and revenue risks to exact commands and pass signals.
Release Gate
Record build, tests, visual QA, secret scan, release asset, revenue claim, and rollback proof.
Control Cost, Models, And Prompts
These templates target the common ways agent workflows leak time, tokens, and repeatable prompts.
Cost Tracker
Track task type, model, estimated tokens, direct cost, human minutes saved, and ROI notes.
Cost Leak Audit
Find leaks from unclear prompts, wrong model choice, weak verification, missing context, and scope drift.
Model Routing Matrix
Choose agent, model, review path, fallback route, escalation rule, and verification proof before the run.
Prompt Regression Test
Compare old and new prompt versions against objective clarity, scope control, verification, and handoff.
Prompt Library
Store reusable prompts with use case, version, change reason, known failure, and eval score.
Workflow Stack Comparison
Compare chat history, plain docs, Notion, spreadsheets, PM tools, and this pack against proof needs.
Handoff, Safety, And Failure Recovery
Use these when work spans agents, reviewers, sessions, or release surfaces.
Agent Handoff
Preserve original goal, current state, files touched, evidence, blockers, and completion requirement.
Scope Freeze
Lock allowed files, do-not-touch paths, commands, stop conditions, and verification rules before edits.
Security Review
Review secrets, personal data, auth, payment, dependency, destructive action, and prompt-leak risk.
Incident Report
Record timeline, impact, missed detection, recovery evidence, and durable prevention rules.
Failure RCA
Capture expected result, actual result, root cause, missed detection, fix, prevention rule, and proof.
Failure Modes Database
Turn repeated mistakes into symptoms, causes, detection rules, fixes, and prevention rules.
Agent Workflow Surfaces
Use these to pick, track, and review Codex, Claude Code, Cursor, and local agent work.
Codex Tracker
Capture repository, goal, allowed files, changed files, verification evidence, blockers, and launch impact.
Claude Code Tracker
Track mode constraints, do-not-touch paths, inputs read, changes, verification, review notes, and blockers.
Cursor Tracker
Record context supplied, files changed, diff summary, verification evidence, review needed, and next action.
Codex vs Claude Code vs Cursor
Pick the right agent owner from proof required, handoff cost, tool access, IDE context, and stop conditions.
Agent Registry
Track each agent, model, tool set, best use, avoid list, prompt contract, verification, and owner.
Operating System Scorecard
Score run capture, scope control, verification, prompt control, cost control, handoff, and revenue proof.
Buyer Proof And Delivery
These public assets keep demand signals separate from proven paid revenue until checkout exists.
Ops Maturity Check
Score run capture, scope control, verification, prompts, costs, failure handling, handoff, and revenue proof before requesting access.
Run Log Builder
Generate one scoped AI-agent run brief, acceptance log row, verification command, and proof packet before requesting the full workspace.
Sample Dashboard
Render the public CSV samples for run log, cost tracker, agent registry, prompt library, verification ledger, and failure modes in one browser view.
Pricing
Review the $29 target price, paid pack contents, checkout status, delivery proof, and revenue evidence rule.
Order Intent Builder
Generate a copy-ready buyer note with preferred checkout, purchase intent, delivery acknowledgement, and the revenue rule.
Interactive ROI Calculator
Estimate value recovered from avoided agent cleanup, then compare it with the $29 target price.
Revenue Proof
Record price, target, checkout URL, receipts, seller dashboard evidence, gross revenue, and non-revenue signals.
Paid Delivery Manifest
Inspect delivered files, ZIP digest, archive test proof, buyer acknowledgement, and support boundary.
Buyer Decision Matrix
Compare cleanup cost, saved time, pack surfaces, buy signals, do-not-buy signals, and preferred checkout.
Checkout Readiness FAQ
Answer buyer objections around checkout status, delivery proof, support, refunds, and revenue evidence.
Buyer Intake Packet
Capture buyer tools, recurring agent problem, cleanup cost, preferred checkout, receipt need, and blockers.
Purchase Approval Brief
Summarize price, business case, expected cleanup time saved, receipt requirement, and delivery method.
Full Pack Request
The paid ZIP is packaged locally and not public. Revenue is counted only when checkout, receipt, payout, or seller-dashboard evidence exists.
Checkout is not live yet. The early-access issue records purchase intent and preferred checkout. At the $29 target price, seven proven paid orders reach $203 gross.